wvs review

    1

    Automated repository auditor for structural analysis, commercial license compliance, and line-level code reviews.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    wvs review

    Example session with this skill installed

    /wvs-review ./src/backend --level 2 --pdf

    • Read your context and instructions
    • Compiled the wvs review
    • Generated the document

    **Analyzing target:** ./src/backend • Level: L2 (structure, commercial) • Output: .review/ ## Repository Review: backend-service | Category | Status/Result | | :--- | :--- | | **Project Name** | backend-service | | **Structure** | Clean, layered architecture detected | | **Tech Stack** | Node.js v20, TypeScript, PostgreSQL | | **License** | AGPL-3.0 (High Risk for commercial use) | | **ISO 5230** | Non-compliant (Missing Open Chain policy) | | **Security Risk** | Grade 4 (High - Exposed secrets in .env.example) | | **Community** | Low activity (Single maintainer) | ### Commercial Risk Audit - **License Warning**: Project uses AGPL-3.0. This requires any derivative works offered over a network to be open sourced under the same terms. - **Security Findings**: Found 3 critical vulnerabilities in sub-dependencies (CVE-2023-xxxx). - **ISO 5230 Gap**: No formal open source program office (OSPO) contact or license manifest found in the root. ### Structural Analysis The repository follows a standard `/src` layout with distinct controllers, services, and repository layers. Testing coverage is present but lacks integration tests for the database layer. **Next steps** 1. Check `REVIEW.md` for the full architectural breakdown. 2. Open `REVIEW-InDepth.md` to see the specific legal mitigations for the AGPL-3.0 license. 3. Use the generated `REVIEW-InDepth.pdf` for stakeholder reporting.

    REVIEW-InDepth.pdf

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Manual code reviews are slow, inconsistent, and often miss critical commercial risks. This skill provides a tiered repository analysis system that scales from structural overviews to deep, line-by-line architectural audits. ## What it does - **Multi-level analysis** performs repository reviews at four distinct depths: structure, commercial/legal, implementation guides, and line-level code review. - **Commercial auditing** evaluates licenses, ISO 5230 compliance, and security risks with 5-stage risk grading. - **Implementation guidance** generates dual-language (Korean/English) onboarding guides specifically tailored for junior developers. - **Granular code review** conducts line-by-line audits adjusted to the reader's technical proficiency level (R0-R4). - **Bulk star analysis** processes entire lists of repositories from markdown files to categorize and rank third-party tools. ## How it works 1. **Initialize** the review by passing a target path and a specific `--level` (1-4) or `--only` flag for specific parts. 2. **Execute sequential parts** starting from inventory and structure, reusing manifest and license data for subsequent commercial or code phases. 3. **Generate artifacts** into the `.review/` directory, saving progress after each part to prevent data loss. 4. **Aggregate results** using the `-O` flag for multi-repo comparisons to identify common vulnerabilities or architectural differences. ## Frameworks & tools This skill integrates with standard repository manifest files (package.json, go.mod, requirements.txt), license files, and project READMEs. It supports `wvs-tools` and `pandoc` for PDF generation and Mermaid for diagram rendering. ## Why this beats prompting it yourself Generic prompts fail to maintain context across massive file trees and often ignore legal or ISO compliance requirements. This skill uses a fixed inventory reuse pattern that ensures the same manifest data informs structural, commercial, and code-level audits without redundant re-scanning. ## Use cases - Audit third-party libraries for ISO 5230 compliance and commercial license risks. - Onboard new developers with automated, dual-language implementation guides. - Compare two or more internal projects to standardize architecture and security. - Perform high-volume competitive analysis on GitHub starred repositories. ## Known limitations Line-by-line code reviews are subject to file and line count limits defined in arguments. PDF generation requires wvs-tools or pandoc installed on the host environment.

    Code review scope is bounded by --max-files and --max-lines-per-file. PDF generation requires wvs-tools or pandoc to be available on the host system.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 5 days ago

    • Free with an Agensi account

    Requires

    Python
    Node.js
    Git

    Permissions

    Network Access
    Read Files
    Write Files
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    Unlike manual prompts, this skill uses a tiered analysis system (Levels 1-4) and an inventory reuse pattern. This ensures that manifest data, license information, and architectural context are consistently applied across structural, legal, and code-level audits without losing context or hitting token limits.

    The skill supports all major package managers including npm (package.json), Go (go.mod), and Python (requirements.txt). It can analyze any repository structure, provided the source files are accessible to your AI agent's environment.

    The skill grades risks on a 5-stage scale specifically designed to meet ISO 5230 standards. It evaluates open-source licenses for commercial compatibility and flags potential legal or security bottlenecks that could affect your project's compliance status.

    Yes, the skill includes a 'Bulk Star Analysis' feature. By providing a markdown file of repository links, the agent will sequentially audit each one, allowing you to rank and categorize third-party tools or perform competitive research at scale.

    While the skill handles the logic and analysis, PDF generation requires 'wvs-tools' or 'pandoc' to be installed on your host system. All other analysis results are generated as markdown and Mermaid diagrams within a local '.review/' directory.

    New in Code Quality & Review